U.S. patent application number 13/736542 was filed with the patent office on 2013-08-22 for method of manufacturing a tissue-engineered prosthesis.
This patent application is currently assigned to TECHNISCHE UNIVERSITEIT EINDHOVEN. The applicant listed for this patent is Frank P.T. BAAIJENS, Carlijn V.C. BOUTEN, Simon P. HOERSTRUP, Anita MOL, Marcel C.M. RUTTEN. Invention is credited to Frank P.T. BAAIJENS, Carlijn V.C. BOUTEN, Simon P. HOERSTRUP, Anita MOL, Marcel C.M. RUTTEN.
Application Number | 20130217128 13/736542 |
Document ID | / |
Family ID | 34933802 |
Filed Date | 2013-08-22 |
United States Patent
Application |
20130217128 |
Kind Code |
A1 |
BOUTEN; Carlijn V.C. ; et
al. |
August 22, 2013 |
METHOD OF MANUFACTURING A TISSUE-ENGINEERED PROSTHESIS
Abstract
Developing heart valves are exposed to dynamic strains by
applying a dynamic pressure difference over the leaflets. The flow
is kept to a minimum, serving only as a perfusion system, supplying
the developing tissue with fresh nutrients. Standard heart valves
were engineered based on B trileaflet scaffolds seeded with cells
isolated from the human saphenous vein. Tissue compaction is
constrained by the stent, inducing increasing pre-strain in the
tissue. The dynamic strains the tissues are exposed to via the
dynamic pressure difference, are estimated using finite element
methods based on the mechanical properties of the neo-tissue, in
order to get inside into the strain distribution over the
leaflet.
Inventors: |
BOUTEN; Carlijn V.C.;
(Eindhoven, NL) ; MOL; Anita; (Eindhoven, NL)
; RUTTEN; Marcel C.M.; (Eindhoven, NL) ;
HOERSTRUP; Simon P.; (Zurich, CH) ; BAAIJENS; Frank
P.T.; (Eindhoven, NL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
BOUTEN; Carlijn V.C.
MOL; Anita
RUTTEN; Marcel C.M.
HOERSTRUP; Simon P.
BAAIJENS; Frank P.T. |
Eindhoven
Eindhoven
Eindhoven
Zurich
Eindhoven |
|
NL
NL
NL
CH
NL |
|
|
Assignee: |
TECHNISCHE UNIVERSITEIT
EINDHOVEN
Eindhoven
NL
Universitaet Zuerich
Zurich
CH
|
Family ID: |
34933802 |
Appl. No.: |
13/736542 |
Filed: |
January 8, 2013 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11816271 |
Feb 15, 2008 |
8399243 |
|
|
PCT/EP2006/000877 |
Feb 1, 2006 |
|
|
|
13736542 |
|
|
|
|
Current U.S.
Class: |
435/402 ;
435/293.1 |
Current CPC
Class: |
C12M 25/14 20130101;
C12N 2521/00 20130101; C12M 35/04 20130101; A61F 2/2412 20130101;
A01N 1/00 20130101; C12M 29/10 20130101; A61K 35/12 20130101; C12N
5/069 20130101; C12N 2533/30 20130101; C12M 21/08 20130101; C12N
5/0691 20130101; A61F 2/2415 20130101 |
Class at
Publication: |
435/402 ;
435/293.1 |
International
Class: |
C12N 5/071 20060101
C12N005/071 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 17, 2005 |
EP |
05003425.5 |
Claims
1.-16. (canceled)
17. A bioreactor for manufacturing a tissue-engineered prosthesis
having at least in an open condition a flow passage, especially
human heart valves, comprising: a bioreactor chamber for inserting
therein a seeded 3D scaffold, at least one perfusion flow means to
provide at least one perfusion flow of a nutrient medium in the
bioreactor chamber, in addition pressuriser means to apply a
dynamic pressure difference over the seeded 3D scaffold and/or
thereon developing tissue to create strain in the 3D scaffold
and/or thereon developing tissue, the pressuriser means having
compressible and decompressive tubing in flow connection with the
bioreactor chamber.
18. The bioreactor of claim 17, wherein at least a portion of the
compressible and decompressible tubing is placed in a cylinder
surrounding the outer surface of said portion of the tubing and
having a port for flow of a compressed fluid, preferably air, into
the cylinder.
19. The bioreactor of claim 18, wherein a magnet valve is provided
being in flow communication with the port of the cylinder to
control the flow of the compressed fluid into the cylinder.
20. The bioreactor of claim 17, wherein a compliance chamber is
provided in flow communication with the bioreactor chamber to
compensate for the displacement of the nutrient medium due to the
pressure difference.
21. The bioreactor of claim 17, wherein the bioreactor chamber has
a first and a second portion each provided with a pressure sensor
and holding means for holding a 3D scaffold and/or the thereon
developing tissue is positioned between the first and second
portion.
22. A method of manufacturing a tissue-engineered prosthesis having
at least in an open condition a flow passage, especially human
heart valves, comprising the steps of: placing a seeded 3D scaffold
in a bioreactor chamber, providing at least one perfusion flow of a
nutrient medium in the bioreactor chamber to supply said seeded 3D
scaffold and/or a thereon developing tissue with nutrients whereby
a flow passage of the seeded 3D scaffold and/or the thereon
developing tissue in relation to the flow passage of the finished
prosthesis is restricted or zero, applying in addition a dynamic
pressure difference over the 3D scaffold and/or thereon developing
tissue depending on the condition and/or stage of tissue
development to create strain in the 3D scaffold and/or thereon
developing tissue, and opening the restricted or zero flow passage
to its finished size.
23. The method of claim 22, wherein the prosthesis is a heart valve
having leaflets and the restriction is achieved by coaptation of
the leaflets.
24. The method of claim 22, wherein the restricted or zero flow
passage is at least 20%, preferably 50%, smaller than the flow
passage of the finished prosthesis.
25. The method of claim 22, wherein the at least one perfusion flow
is less than 50 ml/min, preferably less than 5 ml/min.
26. The method of claim 22, wherein the at least one perfusion flow
has substantially no pulsation.
27. The method of claim 22, wherein the pressure difference is
substantially zero at the beginning, subsequently increasing up to
a mean peak pressure difference and thereafter decreasing to the
end of the duration of stay of the 3D scaffold and/or thereon
developing tissue in the bioreactor chamber.
28. The method of claim 22, wherein a mean peak pressure difference
averaged for 24 hours, is above 25 mmHg, preferably above 45
mmHg.
29. The method of claim 22, wherein the mean pressure difference
over a time period, which is 30-70%, preferably 45-55%, of the
duration of stay in the bioreactor, is substantially zero.
30. The method of claim 29, wherein the mean peak pressure
difference averaged for 24 hours is reached after 60%, preferably
after 70%, of the duration of stay in the bioreactor.
31. The method of claim 30, wherein the modulated pressure
differences per time period, preferably 24 hours, are used to
provide respective strain distribution in that time period.
32. method of claim 22, wherein the dynamic pressure difference has
a frequency of 0.1 to 10 Hz, preferably 1 Hz.
Description
CROSS-REFERENCE TO RELATED PATENT APPLICATION
[0001] This application is a Continuation of U.S. application Ser.
No. 11/816,271, filed Feb. 15, 2008, which is a National Stage
application of PCT/EP2006/000877, filed Feb. 1, 2006, which claims
priority on European Patent Application No. 05003425.5, filed Feb.
17, 2005. The contents of the prior applications are incorporated
by reference in their entirety.
[0002] The invention is directed to a method of manufacturing a
tissue-engineered prosthesis having at least in an open condition a
flow passage, especially human heart valves. The invention is also
directed to a respective bioreactor.
[0003] Such a method is known from WO2004/018008 A1 and
WO2004/101012 A1. The latter also discloses a bioreactor for
producing tissue prosthesis, particularly a heart valve.
[0004] Mechanical conditioning in a bioreactor profoundly affects
the composition and structure and hence mechanical properties of
tissue engineered heart valves. Up to now, the methods and
bioreactors developed for culturing heart valves are flow-based and
mimic the normal heart valve opening and closing behaviour. This
method of conditioning has resulted in enhanced tissue formation
and mechanical properties, however, yet insufficient to serve as
systemic pressure heart valve replacement such as an aortic heart
valve replacement.
[0005] In cardiovascular tissue engineering, stimulation of tissue
formation by mechanical conditioning has proven to be a useful tool
for development of functional cardiovascular structures enabling
growth, repair, and remodeling. The main ways of conditioning a
developing tissue are by applying either flow or strain, or a
combination of both. Various bioreactors have been developed in the
past years to apply specific conditioning protocols to growing
cardiovascular structures.
[0006] Bioreactors that use flow as the main mechanical stimulus
are, for example, the bioreactors developed by Williams and Wick
(2004) and Narita et al. (2004) to engineer blood vessels.
Furthermore, a pulse duplicator system has been developed by
Hoerstrup et al. (2000a,b) to grow heart valves as well as blood
vessels, modified by Sodian et al. (2001, 2002b) to include the
seeding procedure. Bioreactors that use strains as the main
mechanical cue are e.g. the bioreactors developed by Niklason et
al. (1999, 2001) and Seliktar et al. (2000) for tissue engineering
of blood vessels. In these bioreactors, the tissue is exposed to
dynamic strains by applying intraluminal pulsatile pressures via an
inflatable silicone tube. Niklason et al. (1999, 2001) apply
dynamic strains of about 5%, while Seliktar et al. (2000) use
larger strains of about 10%. For myocardial tissue, a comparable
bioreactor is developed by Gonen-Wadmany et al. (2004) applying
dynamic strains (0-12%) to the developing tissue by pulsatile
inflation of a silicone bulb, to which the tissue is attached. The
flow applied to the tissues cultured in these strain-based
bioreactors can be kept to a minimum and are mainly induced by
medium circulation or the movement of the tissue itself. In another
type of bioreactors, the physiological environment of the
cardiovascular structure is mimicked, including both flow and
strain. These bioreactors can be used for testing native tissues,
as well as for tissue engineering and subsequent functionality
testing. A bioreactor in which the physiological pressure
waveforms, present in a blood vessel, can be applied is the system
developed by McCulloch et al. (2004). A bioreactor in which
physiological pressure waveforms can be applied in combination with
physiological flows is developed by Thompson et al. (2002). As
blood vessels are exposed to longitudinal strains in-vivo as well,
Mironov et al. (2003) developed a bioreactor that exposes
developing blood vessels to a physiological environment including
dynamic longitudinal strains. Hildebrand et al. (2004), Dumont et
al. (2002), and Rutten et al. (2005) have developed bioreactors, in
which the exact physiological conditions of a heart valve in-vivo
can be applied, with the latter providing the possibility to
visualize valve function using MRI. Despite these efforts, the
question still remains to what extent the developing tissue should
be exposed to mechanical cues and which mechanical cues are optimal
for tissue development.
[0007] The cells in the tissue engineered structures are
responsible for the formation of the extracellular matrix and via
mechanical conditioning they can be stimulated to produce larger
amounts of extracellular matrix. Apart from the nature and
magnitude of the mechanical cues, the cellular (pheno)type as well
as the culture conditions (either 2D or in a 3D environment) affect
the cellular responses to mechanical loading (Dethlefsen et al.,
1996; Ueba et al., 1997; Watase et al. 1997; Kim et al. 1999;
Chapman et al., 2000; O'Callaghan and Williams, 2000; Kim and
Mooney, 2000; Jockenhoevel et al., 2002; Lee et al., 2002;
Engelmayr et al. 2005). Human saphenous vein cells were chosen as a
cell source for tissue engineering of human heart valve leaflets.
Not only do they represent an easily accessible cell source, they
were also shown to be more sensitive to mechanical cues compared to
human arterial derived cells (Dethlefsen et al. 1996; Schnell et
al., 2001). Although the optimal conditioning procedure for this
specific cell type still has to be determined, increased
extracellular matrix formation has been demonstrated in engineered
valve leaflet tissue-equivalents cultured with these cells using
dynamic strains (Mol et al., 2003). It is, therefore, hypothesized
that the optimal conditioning protocol involves dynamic strains to
stimulate matrix production by the seeded vascular cells. An
initial low shear stress environment combined with exposure of the
developing tissue to physical stimuli has been suggested by Barron
et al. (2003) to be advantageous for initial tissue development.
Shear stresses, by application of flows, will most likely start to
play a significant role at a later stage when the leaflets are
seeded with endothelial cells to provide a non-thrombogenic surface
layer, stabilizing the underlying tissue prior to implantation
(Nackman et al., 1998; Weston and Yoganathan, 2001).
[0008] It is the object of the present invention to improve the
mechanical properties of tissue-engineered prosthesis.
[0009] The object is achieved by a method of manufacturing a
tissue-engineered prosthesis, having at least in an open condition
a flow passage, especially human heart valves, comprising the steps
of: [0010] Placing a seeded 3D scaffold in a bioreactor chamber
[0011] Providing at least one perfusion flow of a nutrient medium
in the bioreactor chamber to supply said seeded 3D scaffold and/or
a thereon developing tissue with nutrients whereby a flow passage
of the seeded 3D scaffold and/or the thereon developing tissue in
relation to the flow passage of the finished prosthesis is
restricted or zero; [0012] Applying in addition a dynamic pressure
difference over the 3D scaffold and/or thereon developing tissue,
depending on the condition and/or stage of tissue development to
create strain in the 3D scaffold and/or thereon developing tissue;
and [0013] opening the restricted or zero flow passage to its
finished size.
[0014] Contrary to the former flow-based methods, the present
invention provides for a strain-based approach. In order to develop
sufficient strain in the 3D scaffold and/or thereon developing
tissue the developing prosthesis is exposed to dynamic strains by
applying a dynamic pressure difference. The strain-based approach
has shown to render tissue-engineered prosthesis with superior
tissue formation and organisation and hence improved mechanical
properties. The appropriate strain can be provided by restricting
or closing the flow passage of the developing prosthesis during the
duration stay in the bioreactor chamber. The restriction or closure
of the flow passage during tissue development is opened to its
final size as usual for the respective prosthesis, especially human
heart valve. For example, only a small opening could be present
allowing only for a minimum perfusion flow of a nutrition medium
from the one side of the prosthesis to the other. In case the flow
passage is completely closed two separate perfusion flows could be
provided, one on each side of the developing prosthesis. With heart
valves the restriction may be achieved by coaptation of the
leaflets.
[0015] The goal of one embodiment is to set up a basic concept to
engineer human heart valve leaflets using dynamic strains in
combination with minimized flows to stimulate extracellular matrix
formation. This is a novel approach as all currently used
conditioning approaches for tissue engineering of heart valves
concentrate on the application of pulsatile flows, with none or
limited tissue straining. For this purpose, a new bioreactor system
is developed, the Diastolic Pulse Duplicator (DPD), to deliver
dynamic strains to the heart valve leaflets by applying a dynamic
pressure difference over the closed leaflets. Requirements for such
a bioreactor system are: 1) simplicity in its use, 2) sized small
to save on culture medium and incubator space, 3) usage of
biocompatible materials, 4) maintenance of sterility over prolonged
periods of time, and 5) the ability to monitor and control the
applied transvalvular pressure. As a stented valve geometry is used
in this study, compaction-induced prestrain develops in the
leaflets as tissue compaction, common in growing and healing
tissues, is constrained by the rigid stent. Medium is circulating
in the DPD at low speed (4 ml/min) to provide oxygen and fresh
nutrients to the developing tissue and to remove waste
products.
[0016] The feasibility of the strain-based approach for human heart
valve tissue engineering is demonstrated. Tissue formation and
mechanical properties of leaflets exposed to dynamic strains in the
DPD and compaction-induced prestrain by the stent were compared
with leaflets exposed to compaction-induced prestrain only. The
latter leaflets were cultured in the DPD with low-speed medium
circulation without application of a dynamic pressure difference.
As controls, unloaded rectangular-shaped valve leaflet
tissue-equivalents, in which compaction was not constrained, were
used. The valve scaffolds were prepared from a non-woven PGA fiber
mesh, coated with P4HB. Fibrin was used as a carrier for the human
saphenous vein cells to ensure homogeneous cell distribution
throughout the scaffold and to render a compact structure suitable
for mechanical conditioning (Mol et al., 2005). The levels of
compaction induced prestrain were estimated, as well as the strain
distribution in the leaflets resulting from the dynamic
transvalvular pressure applied in the DPD, the latter using
numerical analyses.
[0017] In one embodiment the restricted or zero flow passage is at
least 20%, preferably 50% or more, smaller than the flow passage of
the finished prosthesis. A dynamic pressure pulse creates more
strain on the 3D scaffold or developing tissue if a compensation
flow through the flow passage is kept to a minimum or zero. In case
the restriction is formed by means of a coaptation area on the
individual leaflets the flow passage is chosen sufficient to either
frequently of occasionally open the valve.
[0018] In a further variant, the at least one perfusion flow is
less than 50 ml/min, preferably less that 5 ml/min. The perfusion
flow is kept to a minimum and the opening in the 3D scaffold and/or
developing tissue should only be sufficiently sized to allow this
small perfusion flow through it and only having few influence on
the creation of the pressure difference.
[0019] It is of further advantage if the at least one perfusion
flow has substantially no pulsation. Especially the mimic of the
normal flow behaviour through the prosthesis should not be
imitated. In case a roller pump is used, only the influence of the
usual operation of this pump is present and no additional
pulsation. The flow should be kept substantially constant.
[0020] In a further embodiment the pressure difference is
substantially zero at the beginning, subsequently increasing up to
a mean peak pressure difference and thereafter decreasing to the
end of the duration of stay of the 3D scaffold and/or thereon
developing tissue in the bioreactor chamber. First the tissue is
given time to develop without substantial strain. Then the mean
distribution of strain is increased and after reaching a mean peak
pressure difference after a certain time it decreases to the end of
the duration of stay in the bioreactor.
[0021] A mean peak pressure difference, averaged for 24 hours is
above 25 mmHg, preferably above 45 mmHg. This assures that a
significant amount of strain is provided which enhances tissue
formation and organisation.
[0022] Furthermore, the mean pressure difference could be, over a
time period, which is 30-70%, preferably 45-55%, of the duration of
stay in the bioreactor substantially zero. Such a time period gives
tissue sufficient time for development on the 3D scaffold.
[0023] According to another embodiment the mean peak pressure
difference averaged for 24 hours is reached after 60%, preferably
after 70%, of the duration of stay in the bioreactor. As can be
seen, the strain based influence in the tissue formation and
organisation is made in the second half of the duration of stay in
the bioreactor.
[0024] A modulated pressure difference could be used for a specific
time period, preferably 24 hours, to provide a respective strain
distribution in that specific time period. The pressure difference
could almost only be modulated with regard to a mean pressure
difference per time period but also within the respective time
period so as to gain a mean pressure difference.
[0025] The dynamic pressure difference could have a frequency or
0.1-10 Hz, preferably 1 Hz. The dynamic pressure differences have a
frequency, which is similar to an average heart sequence, but not
the flow through the prosthesis is mimicked but the strain in the
prosthesis is provided in this frequency.
[0026] The invention is further directed to a bioreactor for
manufacturing a tissue-engineered prosthesis having at least in an
open condition a flow passage, especially human heart valves,
comprising: [0027] A bioreactor chamber for inserting therein a
seeded 3D scaffold, [0028] at least one perfusion flow means to
provide at least one perfusion flow of a nutritient medium in the
bioreactor chamber, [0029] in addition pressurising means to apply
a dynamic pressure difference over the seeded 3D scaffold and/or
thereon developing tissue to create strain in the 3D scaffold
and/or thereon developing tissue, the pressuriser means having a
compressible and decompressible tubing in flow connection with the
bioreactor chamber.
[0030] The perfusion flow means and the additional pressurising
means are preferably independent from one another so that they
apply a dynamic pressure difference overlaying the at least one
perfusion flow which usually has no pulsation. In order to not have
a contamination of the nutrient medium a pressure pulse from the
outside of the tubing is applied. The tubing is compressible and
returns back to its initial shape after pressure release.
[0031] In a preferred embodiment, at least a portion of the
compressible and decompressible tubing is placed in a cylinder
surrounding the outer surface of said portion of the tubing and
having a port for flow of a compressed fluid, preferably air, into
the cylinder. The cylinder is preferably manufactured from rigid
material so that sufficient pressure can be provided within the
space between the cylinder and the tubing.
[0032] A magnet valve could be provided which is in flow
communication with the port of the cylinder to control the flow of
the compressed fluid into the cylinder. The magnet valve could also
control the outflow of the fluid out of the cylinder. A constant
supply pressure could be provided which is modulated by the magnet
valve so that within the cylinder the predetermined pressure is
present. This allows for an easy and cost efficient modulation of
the pressure.
[0033] Furthermore, a compliance chamber could be provided in flow
communication with the bioreactor chamber to compensate for the
displacement of the nutrient medium due to the pressure difference.
Especially below the developing prosthesis some of the nutrient
medium is displaced. Due to only a minimum of perfusion flow, the
displaced volume of the nutrient volume has to be compensated by
the compliance chamber. After pressure release the displaced
nutrient medium flows back out of the compliance chamber and into
the respective portion of the bioreactor chamber. The overall flow
rate of the nutrition medium is not increased.
[0034] In a further embodiment, the bioreactor chamber has a first
and a second portion each provided with a pressure sensor and
holding means for holding the 3D scaffold and/or the
thereon-developing tissue is positioned between the first and
second portion. With these two sensors the pressure difference from
one side of the developing prosthesis to the other side can be
measured.
[0035] In the following, an embodiment of the present invention is
described in detail. The following description should not, however,
restrict the claims. The claims should be understood in their
broadest meaning. The figures show:
[0036] FIG. 1a,1b: Photographs of the valve scaffold, bottomview
(a) and topview (b).
[0037] FIG. 2a,2b: Schematic drawing of one DPD and its function
(a) and a photograph of six DPD systems in use simultaneously
(b).
[0038] FIG. 3: Finite element mesh of the stented valve geometry.
Because of symmetry only 1/2 of a leaflet is used in the finite
element analyses. This part of the geometry is discretized using
224 hexahedral elements.
[0039] FIG. 4: A graph of the metabolic activity of cells cultured
with the first four changes of the medium that has circulated in
the DPD during the sterility test.
[0040] FIG. 5: A representative transvalvular pressure curve
measured during culturing of the valves.
[0041] FIG. 6a,6b,6c,6d.6e.6f: Representative sections of the valve
leaflet tissues after four weeks of culturing stained with H&E
(a,b,c) and Trichrome Masson (d,e,f).
[0042] FIG. 7a,7b: Stress-strain curves after four weeks of
culturing of non-loaded leaflet tissue-equivalents, leaflets
exposed to prestrain, and leaflets exposed to additional dynamic
strains (a) and the evolution of mechanical properties over time
for dynamically strained leaflets (b).
[0043] FIG. 8a,8b,8c: A graph of the dynamic strain distribution
after four weeks of culturing at the upstream (a) and downstream
(b) surfaces of the leaflets at an applied dynamic transvalvular
pressure difference of 37 mm Hg. The grayscale represents the
amount of dynamic strains in %. The relative amount of the overall
dynamic strains found in the leaflet and the mean value are shown
in the histogram (c).
PREPARATION OF THE LEAFLET TISSUES
The Heart Valve Scaffold
[0044] Trileaflet heart valve scaffolds were fabricated on a
Fastacryl R.degree. stent. The two components, Fastacryl powder and
fluid (PMMA and MMMA, Vertex-dental, the Netherlands) were mixed,
poured into a mold, and allowed to polymerize for 30 minutes. After
complete polymerization, the stent was released from the mold.
Anatomically shaped leaflets, including coaptation areas, were cut
out of non-woven polyglycolic acid meshes (PGA; thickness 1.0 mm;
specific gravity 69 mg/cm.sup.3; Cellon, Luxembourg). The leaflets
were coated with a thin layer of poly-4-hydroxybutyrate (P4HB; MW
1.times.10.sup.6; TEPHA Inc., Cambridge, USA) as described before
(Hoerstrup et al., 2000a). Before evaporation of the solvent, the
leaflets were positioned onto a teflon mold in the shape of a
trileaflet heart valve. The fastacryl stent was placed on top. By
action of the solvent, dissolving the surface layer of the stent,
the leaflets were fixed to the stent. After evaporation of the
solvent, the valve scaffold including the stent was removed from
the mold (FIG. 1a,1b). The valve scaffolds were further dried under
vacuum overnight to remove solvent remnants, after which they were
sterilized using ethylene oxide. The leaflets contain coaptation to
ensure closure of the valve.
Seeding Procedure
[0045] Cells harvested from the human vena saphena magna and
expanded using regular cell culture methods were used (Schnell et
al., 2001). The medium to culture these cells consisted of DMEM
Advanced (Gibco, USA), supplemented with 10% Fetal Bovine Serum
(FBS; PAN Biotech, Germany), 1% GlutaMax (Gibco, USA), and 0.1%
gentamycin (PAN Biotech, Germany). The medium used for seeding and
subsequent tissue culture contained 0.3% gentamycin and additional
L-ascorbic acid 2-phosphate (0.25 mg/ml; Sigma, USA) to promote
extracellular matrix production. The scaffolds were placed in
medium overnight before seeding to facilitate cell attachment by
deposition of proteins. The seeding was performed per leaflet using
fibrin as a cell carrier (Mol et al., 2005). Briefly, the cells
were suspended in a sterile thrombin solution (10 IU/ml medium;
Sigma, USA) in a volume that equals half the void volume of the
scaffold (0.5.times.length.times.width.times.thickness). The cells
in thrombin were mixed with an equal amount of sterile fibrinogen
solution (10 mg actual protein/ml medium; Sigma, USA) and dripped
onto the scaffold. The fibrin solution was taken up by the scaffold
and remained inside due to polymerization of the fibrin gel. The
leaflets were seeded with a density of 4-5 million cells (passage
6-7) per cm2 of scaffold. The seeded valve scaffolds were allowed
to polymerize for 20 minutes in an incubator (37.+-.C and 5% CO2)
before placement into the DPD.
The Diastolic Pulse Duplicator (DPD)
Description of the DPD
[0046] One DPD consists of a bioreactor [A], in which the valve is
cultured, and a medium container [B]. They are connected to each
other via two parallel tubing series [C], which run through a
roller pump [D]. Part of the upper tubing consists of a thicker
tube placed in a polycarbonate cylinder [E]. Compressed air can be
released into this cylinder, via a magnet valve [F], to compress
and decompress the tube, resulting in a pressure difference over
the leaflets. A syringe [G], placed on the bioreactor, serves as a
compliance chamber. The pressures upstream and downstream of the
leaflets is monitored using pressure sensors [H].
[0047] Each DPD consists of two components as shown in FIG. 2 (a):
a bioreactor (height=9 cm, diameter=6 cm), in which the valve is
cultured, and a medium container of similar dimensions, both
fabricated from polycarbonate (KUBRA Kunststoffen, The
Netherlands). The bioreactor itself consists of two parts, the
upper part containing a glass window (Melles Griot BV, The
Netherlands) for visualization of the valve, and a lower part,
which can be screwed together. Silicone rings (van der Heijden, The
Netherlands) were used to seal all components. The bioreactor and
the medium container are connected via two parallel silicone tubing
series (Rubber, the Netherlands). Polypropylene connector parts
(Neolab, Milispec Int., The Netherlands) were used to secure the
tubing. Both tubing series run through a roller pump
(Master.degree.exr, ColeParmer, USA). Part of the upper tubing
serie consists of a thicker silicone tube placed in a polycarbonate
cylinder, with a connection to tubing suitable to withstand air
under pressure (Festo, The Netherlands). This compressed-air tubing
is connected to a compressed-air tap (7 bar). The air pressure is
reduced to 2 bar and runs through a proportional magnet valve
(Festo, The Netherlands) into the polycarbonate cylinder. The
complete DPD is sterilized by ethylene oxide and is placed inside
an incubator, together with the roller pump. Up to six DPDs can be
placed onto one shelf of a normal sized incubator as shown in FIG.
2 (b).
Functioning of the DPD
[0048] About 75 ml of medium is circulating from the medium
container, through the bioreactor, and back into the medium
container, via the roller pump at very low speed (4 ml/min) to
supply fresh nutrients to and to remove waste products from the
developing valve tissue. A sterile filter (0.2 .mu.m, Schleicher
& Schuell Bioscience, Germany) is placed in the lid of the
medium container to oxygenate the circulating medium. About 25 ml
of medium is present in the bioreactor and the tubing series, and
the remaining 50 ml is present in the medium container and can be
changed easily. To apply a dynamic pressure difference over the
leaflets, the silicone tube in the polycarbonate cylinder is
compressed and decompressed by the air coming from the proportional
magnet valve. The transvalvular pressure generated over the valve
leaflets is controlled via a programmable multi-10-card using
LabView software (National Instruments, USA). The shape and
frequency of the pressure wave is programmed for an optimal
tranvalvular pressure. A syringe, connected to the lower part of
the biore actor, filled with 10 ml of medium and 40 ml of air,
serves as compliance chamber. Pressure sensors (BD, Belgium),
connected to both the lower and upper part of the bioreactor, are
used to measure the pressure upstream (at the ventricular side) and
downstream (at the arterial side) of the leaflets. Using the same
multi-10-card and software, the dynamic pressure difference could
be monitored and logged.
Sterility and Biocompatibility
[0049] To prove maintained sterility over several weeks, the DPD
was sterilized by ethylene oxide and filled with 75 ml of culture
medium without the addition of gentamycin. The system was fully
functional for a period of three weeks. During the testing period,
a culture flask filled with culture medium, without gentamycin,
served as control. The medium in the medium container was changed
every 3-4 days as well as two thirds of the medium in the culture
flask. The medium was checked microscopically for contamination and
was stored at -20.degree. C. until further use.
[0050] In order to ensure that the materials used in the DPD are
suitable for cultivation of heart valves and that no toxic
components are being released, a biocompatibility test was
performed using the medium that was stored during the sterility
test. Human saphenous vein cells were seeded into 24 wells plates
(30.000 cells/well) and left overnight to attach and spread in an
incubator in normal cell culture medium. The next day, the normal
cell culture medium was replaced by the medium of the sterility
test that had circulated through the DPD and the control medium
(five wells per test group). The medium stored after the first four
medium changes, as well as the control medium from the culture
flask, was used. The cells were allowed to grow for three
subsequent days, after which the metabolic activity of the cells,
as a measure for viability, was determined using an MTT test
(Mosmann, 1983). Briefly, MTT (Sigma, USA) in solution (5 mg/ml in
PBS) was diluted in medium and added to the cells. After one hour
of incubation at 37.+-.C and 5% CO2, metabolic active cells had
converted the MTT salt into purple crystals located inside the
cells. Isopropanol (VWR International, USA), containing 10% formic
acid (Sigma, USA) was added to the cells to release and dissolve
the purple crystals and the optical density of the solution was
determined. The results are expressed as percentages with respect
to the control, which was set at 100%.
Tissue Culture and Mechanical Conditioning
[0051] Three experimental groups of valve leaflet tissues were
engineered to compare tissue formation and to show the feasibility
of the strain-based approach. The first group comprises rectangular
shaped valve leaflet tissue-equivalents. They were cultured
statically in a cell culture flask for up to four weeks and could
compact freely, serving as non-loaded controls. The second and
third group consisted of valve leaflets, all cultured in the DPD.
Directly after placement in the DPD, all leaflets were exposed to
medium circulation at low speed (4 ml/min) to supply fresh
nutrients to the developing tissue. Valve leaflets were cultured up
to four weeks, exposed to continuous medium circulation and
prestrain due to compaction constrained by the stent (group 2).
Additional valve leaflets were exposed to dynamic strains at 1 Hz
in addition to the prestrain and continuous medium circulation
(group 3). The pressures upstream and downstream of the valves, as
well as the dynamic transvalvular pressure, were recorded every
three hours. The valve leaflets of group 3 were cultured up to two,
three and four weeks. The medium in the medium container, as well
as two thirds of the medium in the culture flask, for group 1, was
replaced every three to four days.
Evaluation of Tissue Formation
Qualitative Tissue Analysis
[0052] Tissue formation in all groups was analyzed by histology
after four weeks. Representative pieces were fixed in
phosphate-buffered formalin (Fluke, USA) and embedded in paraffin.
Sections were cut at 5 .sup.1m thickness and studied by
Haematoxylin and Eosin (H&E) staining for general tissue
morphology and Trichrome Masson staining for collagen formation, as
collagen is the main load-bearing component of the extracellular
matrix.
Mechanical Testing
[0053] The mechanical properties of circumferential strips of the
engineered leaflet tissues of the three groups after four weeks of
culturing were determined by uniaxial tensile tests. As thickness
measurements of the fresh strips are practically difficult, the
thickness was determined from representative histology sections.
Stress-strain curves were obtained using an uniaxial tensile tester
(Instron, Belgium, model 4411, equipped with a load cell of 10 N)
with a constant strain rate of 1.7% per second. To get insight into
the evolution of mechanical properties in time, the mechanical
properties of circumferential strips of dynamically strained
leaflets after two and three weeks of culturing were additionally
determined.
Estimation of Strains in the Leaflets
[0054] To estimate the amount of prestrain in the leaflets, due to
compaction constrained by the stent, the difference in size of the
leaflets in the stent after culturing and after release from the
stent was determined. The leaflets, having an initially bulged
shape in the stent, did straighten during culturing due to free
compaction of the neo-tissue. The initial circumferential size of
the leaflets was 23 mm, whereas in fully straightened leaflets this
size was reduced to 20 mm, indicating that the leaflets could
compact freely up to 13% during culturing. The circumferential size
of the leaflets was measured after release of the leaflets from the
stent of the valves exposed to prestrain and medium circulation
after four weeks. This measured size was divided by the maximally
straightened size of 20 mm to calculate the amount of prestrain in
the leaflets. This value was subsequently extracted from 1 and
multiplied by 100% to obtain the amount expressed in a percentage
of prestrain.
[0055] Finite element analyses were used to estimate the amount and
distribution of the dynamic strains in the leaflets resulting from
the applied dynamic transvalvular pressure. The finite element mesh
of the stented valve geometry is shown in FIG. 3. The configuration
shown in this figure was assumed to be stress-free and because of
symmetry only 1/2 of a valve leaflet was used in the finite element
analyses. At the symmetry edge, nodal displacements in the normal
direction were suppressed, whereas at the fixed edge all nodal
displacements were set to zero. At the free edge, a contact surface
was defined to model coaptation of adjacent leaflets. The
transvalvular pressure ptv was subsequently applied to the
downstream surface of the leaflets. In this study, the finite
element package SEPRAN was used (Segal, 1984).
[0056] The valve tissues were modeled as an incompressible
generalized Neo-Hookean material:
.sigma.=-pl+G(B-l) (1.1)
with .sigma. the Cauchy stress, p the hydrostatic pressure, l the
unity tensor and G the shear modulus of the material. The left
Cauchy-Green deformation tensor was defined as B=FF.sup.T with F
the deformation gradient tensor. To describe the (potential)
non-linear behavior of the leaflet tissues, the following
expression for G was used:
G=G0(l.sub.1(B)/3).sup.n (1.2)
with G0 and n material parameters and l.sub.1(B)=trace(B) the first
invariant of B. The parameter of n is used to control the degree of
non-linearity: n>0 results in strain hardening, n<0 in strain
stiffening and for n=0 the classical Neo-Hookean model is obtained.
The material parameters were obtained by fitting the constitutive
law (Eq. 1.1) to the mean results before failure of the uniaxial
tensile tests of the leaflet tissues exposed to dynamic strains
after two, three, and four weeks of culturing.
Results
The DPD
[0057] The tested medium (dark gray bars) was compared to control
medium (light gray bars), which were set at 100%. The metabolic
activity was in all cases 80-100% compared to the controls and,
therefore, the DPD can be considered as being biocompatible. The
error bars represent the standard error of the mean.
[0058] The medium, without addition of antibiotics, that had
circulated in the DPD did not show any macroscopic or microscopic
signs of contamination for the complete test period. The system was
easy to handle and the risk of contamination during medium
replacement was minimal. The results of the biocompatibility test,
performed for the first four medium changes during the sterility
test, are shown in FIG. 4. The metabolic activity of the cells
cultured with the four tested media were all within 80 to 100% when
compared to the controls, indicating the DPD to be biocompatible
and suitable for cultivation of human heart valves.
Dynamic Straining Protocol
[0059] A representative transvalvular pressure curve, measured
during culturing is shown in FIG. 5. The definitions of the DC
offset, the cyclic transvalvular pressure difference, and the
dynamic transvalvular pressure difference are represented. The
permanent transvalvular pressure, present over the valve leaflets,
is referred to as the DC offset, which is the average value of the
difference in pressure upstream and downstream of the valve. The
cyclic transvalvular pressure is defined as the peak-to-peak
transvalvular pressure value. The maximum trans-valvular pressure
value, including the DC offset and the cyclic transvalvular
pressure, is referred to as the applied dynamic transvalvular
pressure.
[0060] The dynamic transvalvular pressure was averaged for each
day, resulting in dynamic trans-valvular pressures increasing
gradually from 0 to about 80 mm Hg within the first two weeks of
culturing. The last two weeks of culturing, the pressures were
lowered, due to the expected loss of the support function of the
scaffold, and kept constant at about 37 mm Hg.
Evaluation of Tissue Formation
Histology
[0061] Sections stained with H&E and Trichrome Masson (FIG. 6)
showed superior tissue formation in the leaflets (FIG. 6(a), 6(b),
6(d), 6(e)) when compared to the non-loaded rectangular leaflet
tissue-equivalents (FIG. 6(c), 6(f)). The tissue of the leaflets
cultured with additional dynamic strains (FIG. 6(a), 5.6(d))
appeared to be more homogeneous and denser when compared to the
leaflets exposed to prestrain only (FIG. 6(b), 6(e)). Collagen
could be identified in the leaflets, either cultured using
prestrain or additional dynamic strains, after four weeks of
culturing (FIG. 6(e), 6(d)). The bars in the images represent
scales of 350 .sup.1m. The dynamically strained leaflets (a,d) and
the leaflets exposed to prestrain only (b,e) showed superior tissue
development when compared to the non-loaded valve leaflet
tissue-equivalents (c,f). The dynamically strained leaflet tissue
seemed more homogeneous and denser as compared to the leaflets
exposed to prestrain only. Collagen, stained blue, could be
identified in the leaflets (d,e).
Mechanical Tests
[0062] Representative stress-strain curves for all groups after
four weeks of culturing are shown in FIG. 7(a). The non-loaded
leaflet equivalents showed linear behavior, while the leaflets,
exposed to compaction-induced prestrain and either with or without
additional dynamic strains, showed more tissue-like non-linear
behavior after four weeks of culturing (a) The evolution of
mechanical properties with increasing culture time is shown in FIG.
7(b) for the leaflets exposed to dynamic strains. After two weeks
of culturing, the mechanical behavior was linear, representing
scaffold-like behavior. After three and four weeks, the tissue
showed more non-linear mechanical behavior, representative for
tissue contribution.
Estimation of Strains in the Leaflets
Prestrain
[0063] The amount of prestrain in the leaflets of the valves
exposed to prestrain only after four weeks of culturing, assuming
maximal straightening of the leaflets, was shown to vary between 3
and 5%.
Dynamic Strains
[0064] The constitutive law (Eq. 1.1) fitted the results of the
uniaxial tensile tests quite reasonable. Table 1.1 summarizes the
input parameters for the finite element model, the mean absolute
error of the fits, and the estimated mean dynamic strains after
two, three and four weeks of culturing, based on a transvalvular
pressure difference of 37 mm Hg. The mean dynamic strains in the
leaflets increased from 8% after two weeks to about 20% after four
weeks of culturing. The dynamic strain distribution in the leaflets
after four weeks of culturing is shown in FIG. 8a,8b,8c, with in
FIGS. 8 (a) and 8(b) the strain distribution within one leaflet at
respectively the upstream and downstream surface and in FIG. 8(c)
the estimated range of overall dynamic strains within a
leaflet.
TABLE-US-00001 TABLE 1.1 Summary of the input parameters for the
finite element analyses of the dynamically strained leaflets after
two, three, and four weeks of culturing. The parameter n has been
set to zero for the leaflets after two weeks of culturing, due to
the observed linear behavior. Furthermore, the mean absolute errors
of the fits of the constitutive law (Eq. 1.1) to the results of the
uniaxial tensile tests and the resulting estimated mean dynamic
strains are represented. Mean Culture time Thickness G.sub.0 n
absolute error Mean dynamic [weeks] [mm] [kPa] [--] of fit [kPa]
strain [%] 2 0.80 188 0.0 0.95 8 3 0.59 51 8.6 3.93 24 4 0.63 59
9.2 4.61 20
DISCUSSION
[0065] Mechanical stimulation of tissue formation is a well-known
technique in tissue engineering of cardiovascular structures to
improve tissue formation and organization. Various conditioning
approaches are being employed in bioreactor systems, from
flow-based to strain-based to even mimicking the exact
physiological environment in the body. For tissue engineering of
blood vessels, flow-based as well as strain-based approaches are
used, however, for tissue engineering of heart valves, only
flow-based approaches have been described. The optimal conditioning
protocol depends on several factors, such as the sensitivity of the
cell pheno-type and source (i.e. animal or human) to mechanical
cues, the scaffold used, the transfer of the mechanical cues from
the scaffold to the cells, and the magnitude and type of mechanical
cue.
[0066] If the forms was put on tissue engineering of human heart
valve leaflets engineered from human saphenous vein cells seeded
using fibrin as a cell carrier onto PGA/P4HB trileaflet heart valve
scaffolds (FIG. 1a,1b). For this particular cell type, being more
sensitive to mechanical stimulation as compared to human arterial
derived cells, we previously have shown a large impact of cyclic
straining during culturing on tissue formation (Mol et al., 2003).
In this former study, the amount of extracellular matrix formation
increased when larger strains were used. Flow is most likely not
necessitated in the early phase of tissue development, but might
start to play a significant role as soon as the leaflets are to be
seeded with endothelial cells in a later phase. Flow might then
stabilize the tissue via signaling by the endothelial cells and
prepare the tissue for subsequent implantation. This study
describes a strain-based approach to tissue engineer human heart
valve leaflets. In order to expose the developing heart valve
leaflets to dynamic strains, a pressure difference had to be
applied over the leaflets, mimicking the diastolic phase in the
heart. A novel bioreactor, the Diastolic Pulse Duplicator (FIG.
2a,2b), is developed for this purpose, which can expose the
developing tissue to increasing amounts of dynamic strains. Besides
application of dynamic strains, the leaflets cultured in this study
were exposed to prestrain, due to tissue compaction constrained by
the stented geometry. The flow was kept low (4 ml/min) and served
solely to provide the tissue with sufficient fresh nutrients and to
remove waste products. The DPD was very easy to handle, sized small
with a total medium volume of only 75 ml per valve, proven to be
biocompatible (FIG. 4), and sterility could be maintained over
prolonged periods of time.
[0067] Rectangular non-loaded valve leaflet tissue-equivalents were
shown to render much less tissue formation after four weeks of
culturing when compared to the superior tissue formation in the
leaflets, exposed to either prestrain only as well as to additional
dynamic strains (FIG. 6). The tissue of the leaflets exposed to
dynamic strains appeared to be more homogeneous and denser packed
as compared to leaflets exposed to prestrain only, however, this
concerned only qualitative observations by histology. The
mechanical properties (FIG. 7a,7b) of the leaflets exposed to
dynamic strains showed increased non-linear tissue-like behavior
over time, indicating increasing amounts of tissue and collagen.
After four weeks of culturing, the non-loaded rectangular shaped
leaflet tissue-equivalents showed linear behavior, representative
for scaffold behavior. The leaflets, exposed to prestrain alone as
well as to additional dynamic strains showed non-linear behavior
after four weeks of culturing, correlating with the larger amounts
of tissue found in the leaflets as compared to the non-loaded
leaflet tissue-equivalents. The continuous medium circulation in
the DPD will most likely contribute to the improved tissue
formation in the leaflets as well.
[0068] Prestrain alone obviously resulted in abundant amounts of
tissue. The prestrain is estimated to be in the range of 3 to 5%
and might already be sufficient for optimized tissue formation.
However, dynamic straining might further enhance tissue
organization and, furthermore, might represent a valuable tool to
maintain the bulged shape of the leaflets by some deformation of
the scaffold and the tissue. Furthermore, for future applications
towards tissue engineering of stentless human heart valves, dynamic
strains will be of larger importance as the prestrains in stentless
valves will be much less.
[0069] To get insight into the amount of dynamic strains applied to
the leaflets, the dynamic strain distribution was estimated. The
material properties of the leaflets determine the amount of dynamic
strains at a given dynamic transvalvular pressure. The
transvalvular pressure was monitored in the DPD during culturing
(FIG. 5). Theoretically, the amount of dynamic strains can be
monitored using markers on the leaflets and subsequent imaging of
the leaflets being strained, followed by image analyses. As this is
a time-consuming and in practice rather difficult method, we have
chosen to estimate the dynamic strains in the leaflets by finite
element analyses. The material parameters of the developing
leaflets, determined by fitting the constitutive law (Eq. 1.1) to
the mean results of the uniaxial tensile tests before failure, were
used to serve as input for the finite element analyses. The mean
dynamic strain in the valve leaflets cultured in this study, as an
example for future use, were estimated to vary from 8% at two weeks
to 20% at four weeks of culturing (FIG. 8a,8b,8c and Table 1.1).
The estimated dynamic strains increased between week two and three,
while the applied dynamic transvalvular pressures were similar,
indicating loss of scaffold support, resulting in a decreased
stiffness of the tissues, as shown by a lower shear modulus after
three weeks of culturing. For the stented valve geometry, the
amount of applied dynamic strains in this study might be too large
for optimized tissue formation, when taking the prestrain into
account, but application of large dynamic strains in the DPD was
shown feasible. As currently the strains in the leaflets are
determined afterwards, a future research focus for optimization of
the Diastolic Pulse Duplicator is the development of a non-invasive
method to determine tissue strains directly during culturing and to
integrate this feature in a feedback loop to control the magnitude
of the strains.
[0070] In the finite element analysis, an initial stress-free
configuration of the leaflets is assumed, which is not completely
true as the valve leaflets show compaction during culturing. The
material properties of the leaflets are assumed to be homogeneous
and isotropic, which might not be the case after a certain culture
period due to the influence of prestrain in mainly the
circumferential direction and the local dynamic strain distribution
in the leaflets. Therefore, leaflets to be cultured in the DPD will
be mechanically tested in both circumferential and radial direction
to identify possible anisotropic properties. In case of anisotropic
properties, an extended model should be used for the finite element
analyses as described by Driessen et al. (2004).
REFERENCES
[0071] Barron, V., Lyons, E., Stenson-Cox, C., McHugh, P. E, and
Pandit, A. (2003). Bioreactors for cardiovascular cell and tissue
growth: a review. Annals of biomedical engineering, 31, 1017-1030.
[0072] Chapman, G. B., Durante, W., Hellums, J. D., and Schafer, A.
L (2000). Physiological cyclic stretch causes cell cycle arrest in
cultured vascular smooth muscle cells. American journal of
physiology. Heart and circulatory physiology, 278, H748-H754.
[0073] Dethlefsen, S. M., Shepro, D., and D'Amore, P. A. (1998).
Comparison of the effects of mechanical stimulation on venous and
arterial smooth muscle cells in vitro. Journal of vascular
research, 33, 405-413. [0074] Driessen, N. J. B., Bouten, C. V. C.,
and Baaijens, F. P. T. (2004). A structural constitutive model for
collagenous cardiovascular tissue incorporating the angular fiber
distribution. Journal of biomechanical engineering, [0075] Dumont,
K., Yperman, J., Verbeken, E., Segers, P., Meuris, B.,
Vandenberghe, S., Flameng, W., and Verdonck, P. R. (2002). Design
of a new pulsatile bioreactor for tissue engineered aortic heart
valve formation. Artificial organs, 26, 710-714. [0076] Engelmayr,
G. C., Rabkin, E., Sutherland, F. W. H., Schoen, F. J., Mayer, J.
E., and Sacks, M. S. (2005). The independent role of cyclic flexure
in the early in vitro development of an engineered heart valve
tissue. Biomaterials, 26, 175-187. [0077] Gonen-Wadmany, M.,
Gepstein, L., and Seliktar, D. (2004). Controlling the cellular
organization of tissue engineered cardiac constructs. Annals of the
New York academy of sciences, 1015, 299-311. [0078] Hildebrand, D.
K., Wu, Z. J., Mayer, J. E., and Sacks, M. S. (2004). Design and
hydrodynamic evaluation of a novel pulsatile bioreactor for
biologically active heart valves. Annals of biomedical engineering,
32, 1039-1049. [0079] Hoerstrup, S. P., Sodian, R., Daebritz, S.,
Wang, J., Bacha, E. A., Martin, D. P., Moran, A. M., Guleresian,
J., Sperling, J. S., Kaushal, S., Vacanti, J. P., Schoen, F. J.,
and Mayer, J. E. (2000a). Functional living trileaflet heart valves
grown in vitro. Circulation, 102 (suppl III), 11149-11149. [0080]
Hoerstrup, S. P., Kadner, A., Breymann, C., Maurus, C. F., Guenter,
C. I., Sodian, R., Visjager, J. F., Zund, G., and Turina, M. I.
(2002a). Living, autologous pulmonary artery conduits tissue
engineered from human umbilical cord cells. Annals of thoracic
surgery, 74, 46-52. [0081] Hoerstrup, S. P., Zund, G., Cheng, S.,
Melnitchouk, S., Kadner, A., Sodian, R., Kolb, S. A., and Turina,
M. (2002b). A new approach to completely autologous cardiovascular
tissue in humans. ASAIO journal, 48, 234-238. [0082] Jockenhoevel,
S., Zund, G., Hoerstrup, S. P., Schnell, A., and Turina, M. (2002).
Cardiovascular tissue engineering: a new laminar flow chamber for
in vitro improvement of mechanical tissue properties. ASAIO
journal, 48, 8-11. [0083] Kim, B., Nikolovski, J., Bonadio, J., and
Mooney, D. J. (1999). Cyclic mechanical strain regulates the
development of engineered smooth muscle tissue. Nature
biotechnology, 17, 979-983. [0084] Kim, B. and Mooney, D. J.
(2000). Scaffolds for engineering smooth muscle under cyclic
mechanical strain conditions. Journal of biomechanical engineering,
122, 210-215. [0085] Lee, A. A., Graham, D. A., Cruz, S. D.,
Ratcliffe, A., and Karlon, W. J. (2002). Fluid shear stress-induced
alignment of cultured smooth muscle cells. Journal of biomechanical
engineering, 124, 37-43. [0086] McCulloch, A. D., Harris, A. B.,
Sarraf, C. E., and Eastwood, M. (2004). New multi-cue bioreactor
for tissue engineering of tubular cardiovascular samples under
physiological conditions. Tissue engineering, 10, 565-573. [0087]
Mironov, V., Kasayanov, V., McAllister, K., Oliver, S., Sistino,
J., and Markwald, R. (2003). Perfusion bioreactor for vascular
tissue engineering with capacities for longitudinal stretch.
Journal of craniofacial surgery, 14, 340-347. [0088] Mol, A.,
Bouten, C. V. C., Zund, G., Guenter, C. I., Visjager, J. F.,
Turina, M., Baaijens, F. P. T., and Hoerstrup, S. P. (2003). The
relevance of large strains in functional tissue engineering of
heart valves. Thoracic and cardiovascular surgeon, 51, 78-83.
[0089] Mol, A., van Lieshout, M. I., dam de Veen, G. C.,
Neuenschwander, S., Hoerstrup, S. P., Baaijens, F. P. T., and
Bouten, C. V. C. (2005). Fibrin as a cell carrier in cardiovascular
tissue engineering applications. Biomaterials, 26, 3113-3121.
[0090] Mosmann, T. (1983). Rapid colorimetric assay for cellular
growth and survival: application to proliferation and cytotoxicity
assays. Journal of immunological methods, 65, 55-63. [0091]
Nackman, G. B., Fillinger, M. F., Shafritz, R., Wei, T., and
Graham, A. M. (1998). Flow modulates endothelial regulation of
smooth muscle cell proliferation: a new model. Surgery, 124,
353-361. [0092] Narita, Y., Hata, K., H, K., Usui, A., Ueda, M.,
and Ueda, Y. (2004). Novel pulse duplicating bioreactor system for
tissue-engineered vascular constructs. Tissue engineering, 10,
1224-1233. [0093] Niklason, L. E., Gao, J., Abbott, W. M., Hirschi,
K. K., Houser, S., Marini, R., and Langer, R. (1999). Functional
arteries grown in vitro. Science, 284, 489-493. [0094] Niklason, L.
E., Abott, W., Gao, J., Klagges, B., Hirschi, K. K., Ulubayram, K.,
Conroy, N., Jones, R., Vasanawala, A., Sanzgiri, S., and Langer, R.
(2001). Morphological and mechanical characteristics of engineered
bovine arteries. Journal of vascular surgery, 33, 628-638. [0095]
O'Callaghan, C. J. and Williams, B. (2000). Mechanical strain
induced extracellular matrix production by human vascular smooth
muscle cells: Role of tgf-beta. Hypertension, 36, 319-324. [0096]
Rutten, M. C. M., Wijlaars, M. W., Mol, A., van Dam, E. A.,
Strijkers, G. J., Nicolay, K., and Baaijens, F. P. T. (2005). The
valve exerciser: a novel bioreactor for physiological loading of
tissue-engineered aortic valves. Journal of biomechanics. [0097]
Schnell, A. M., Hoerstrup, S. P., Zund, G., Kolb, S., Sodian, R.,
Visjager, J. F., Grunenfelder, J., Suter, A., and Turina, M.
(2001). Optimal cell source for tissue engineering: Venous vs.
aortic human myofibroblasts. Thoracic and cardiovascular surgeon,
49, 221-225. [0098] Segal, A. (1984). SEPRAN user manual, standard
problems and programmers guide. Leidschendam: Ingenieursbureau
SEPRA, the Netherlands. [0099] Seliktar, D., Black, R. A., Vito, R.
P., and Nerem, R. M. (2000). Dynamic mechanical conditioning of
collagen-gel blood vessel constructs induces remodeling in vitro.
Annals of biomedical engineering, 28, 351-362. [0100] Sodian, R.,
Lemke, T., Loebe, M., Hoerstrup, S. P., Potapov, E. V., Hausmann,
H., Meyer, R., and Hetzer, R. (2001). New pulsatile bioreactor for
fabrication of tissue-engineered patches. Journal of biomedical
materials research, 58, 401-405.
[0101] Sodian, R., Lemke, T., Fritsche, C., Hoerstrup, S. P., Fu,
P., Potapov, E. V., Hausmann, H., and Hetzer, R. (2002b).
Tissue-engineering bioreactors: a new combined cell-seeding and
perfusion system for vascular tissue engineering. Tissue
engineering, 8, 863-870. [0102] Thompson, C. A., Colon-Hernandez,
P., Pomerantseva, I., MacNeil, B. D., Nasseri, B., Vacanti, J. P.,
and Oesterle, S. N. (2002). A novel pulsatile, laminar flow
bioreactor for the development of tissue engineered vascular
structures. Tissue engineering, 8, 1083-1088. [0103] Ueba, H.,
Kawakami, M., and Yaginuma, T. (1997). Shear stress as an inhibitor
of vascular smooth muscle cell proliferation: Role of transforming
growth factor-beta1 and tissue-type plasminogen activator.
Arteriosclerosis, thrombosis, and vascular biology, 17, 1512-1516.
[0104] Watase, M., Awolesi, M. A., Ricotta, J., and Sumpio, B. E.
(1997). Effects of pressure on cultured smooth muscle cells. Life
sciences, 61, 987-996.
[0105] Weston, M. W. and Yoganathan, A. P. (2001). Biosynthetic
activity in heart valve leaflets in response to in vitro flow
environments. Annals of biomedical engineering, 29, 752-763. [0106]
Williams, C. and Wick, T. M. (2004). Perfusion bioreactor for small
diameter tissue-engineered arteries. Tissue engineering, 10,
930-941.
* * * * *